STRUCTURED CLINICAL DOCUMENTATION FOR PATIENT CARE AND PRACTICE-BASED RESEARCH IN MCI AND DEMENTIA

Abstract

To improve quality of care of patients with mild cognitive impairment and dementia, we used the electronic medical record to develop structured clinical documentation support (SCDS) tools following quality guidelines for MCI and dementia. The tools also prompted enrollment in practice-based research, including in a DNA biobank beginning in September 2014. The toolkits assess cognitive, behavioral/psychological, and motor symptoms and also include the Barthel Index, Functional Activities Questionnaire (FAQ), Geriatric Depression Scale (GDS), and Montreal Cognitive Assessment (MoCA). The Functional Assessment Stage Test is used for disease staging. All patients referred to the NorthShore University Health System Department of Neurology for an evaluation of changes in cognition or behavior are evaluated with our SCDS toolkit at initial visit and annual follow-up visits. At the initial visit, a diagnosis of MCI or dementia electronically prompted the physician to enroll the patient in the DNA biobank. There were 251 patients (103 with MCI and 148 with dementia) enrolled in the biobank. Median age at enrollment was 76 (mean 74.6). Most patients were initially evaluated within 4 years of symptom onset, and a similar age at onset was noted in men and women. Executive symptoms and irritability appeared to occur more frequently in men compared to women but did not reach statistical significance. Overall, there was no significant gender difference in current symptoms at the initial evaluation. However, there was a statistically significant difference between men and women for the place of living (Fisher test, p-value=0.001), with more women living in assisted living and more men living at home, which may reflect women outliving their spouses. There was no gender difference in MoCA scores or in the distribution of MCI or dementia diagnoses. Pairwise correlations found inverse correlations between age at onset and disease duration and between FAQ and MoCA. The GDS and MoCA were directly correlated. Multidimensional comparisons using principal component analysis revealed that the FAQ accounted for most of the variance between the 6 continuous trait measures (age at onset, disease duration, Barthel Index, FAQ, MoCA, and GDS). Our SCDS toolkit efficiently measures and improves the quality of care in memory disorders and through data capture has the potential to support multicenter practice based research.